scholarly journals Controlling Fairness and Bias in Dynamic Learning-to-Rank (Extended Abstract)

2021 ◽  
Author(s):  
Marco Morik ◽  
Ashudeep Singh ◽  
Jessica Hong ◽  
Thorsten Joachims
Keyword(s):  
Theoretical Foundation ◽  
Learning Algorithm ◽  
Learning To Rank ◽  
Ranking Function ◽  
Implicit Feedback ◽  
Music Video ◽  
Dynamic Learning ◽  
Unbiased Estimators ◽  
Online Platforms ◽  
Feedback Data

Rankings are the primary interface through which many online platforms match users to items (e.g. news, products, music, video). In these two-sided markets, not only do the users draw utility from the rankings, but the rankings also determine the utility (e.g. exposure, revenue) for the item providers (e.g. publishers, sellers, artists, studios). It has already been noted that myopically optimizing utility to the users -- as done by virtually all learning-to-rank algorithms -- can be unfair to the item providers. We, therefore, present a learning-to-rank approach for explicitly enforcing merit-based fairness guarantees to groups of items (e.g. articles by the same publisher, tracks by the same artist). In particular, we propose a learning algorithm that ensures notions of amortized group fairness, while simultaneously learning the ranking function from implicit feedback data. The algorithm takes the form of a controller that integrates unbiased estimators for both fairness and utility, dynamically adapting both as more data becomes available. In addition to its rigorous theoretical foundation and convergence guarantees, we find empirically that the algorithm is highly practical and robust.

scholarly journals Fast Adaptively Weighted Matrix Factorization for Recommendation with Implicit Feedback

2020 ◽  
Vol 34 (04) ◽  
pp. 3470-3477
Author(s):  
Jiawei Chen ◽  
Can Wang ◽  
Sheng Zhou ◽  
Qihao Shi ◽  
Jingbang Chen ◽  
...  
Keyword(s):  
Matrix Factorization ◽  
Learning Algorithm ◽  
Implicit Feedback ◽  
Model Learning ◽  
Adaptive Weights ◽  
Network Function ◽  
Fast Learning ◽  
Weighted Matrix ◽  
Real World Datasets ◽  
Feedback Data

Recommendation from implicit feedback is a highly challenging task due to the lack of the reliable observed negative data. A popular and effective approach for implicit recommendation is to treat unobserved data as negative but downweight their confidence. Naturally, how to assign confidence weights and how to handle the large number of the unobserved data are two key problems for implicit recommendation models. However, existing methods either pursuit fast learning by manually assigning simple confidence weights, which lacks flexibility and may create empirical bias in evaluating user's preference; or adaptively infer personalized confidence weights but suffer from low efficiency.To achieve both adaptive weights assignment and efficient model learning, we propose a fast adaptively weighted matrix factorization (FAWMF) based on variational auto-encoder. The personalized data confidence weights are adaptively assigned with a parameterized neural network (function) and the network can be inferred from the data. Further, to support fast and stable learning of FAWMF, a new specific batch-based learning algorithm fBGD has been developed, which trains on all feedback data but its complexity is linear to the number of observed data. Extensive experiments on real-world datasets demonstrate the superiority of the proposed FAWMF and its learning algorithm fBGD.

scholarly journals Improving Implicit Recommender Systems with View Data

2018 ◽  
Author(s):  
Jingtao Ding ◽  
Guanghui Yu ◽  
Xiangnan He ◽  
Yuhan Quan ◽  
Yong Li ◽  
...  
Keyword(s):  
Recommender Systems ◽  
Matrix Factorization ◽  
Time Complexity ◽  
Learning Algorithm ◽  
State Of The Art ◽  
Implicit Feedback ◽  
Model Parameters ◽  
New Learning ◽  
Real World Datasets ◽  
Feedback Data

Most existing recommender systems leverage the primary feedback data only, such as the purchase records in E-commerce. In this work, we additionally integrate view data into implicit feedback based recommender systems (dubbed as Implicit Recommender Systems). We propose to model the pairwise ranking relations among purchased, viewed, and non-viewed interactions, being more effective and flexible than typical pointwise matrix factorization (MF) methods. However, such a pairwise formulation poses efficiency challenges in learning the model. To address this problem, we design a new learning algorithm based on the element-wise Alternating Least Squares (eALS) learner. Notably, our algorithm can efficiently learn model parameters from the whole user-item matrix (including all missing data), with a rather low time complexity that is dependent on the observed data only. Extensive experiments on two real-world datasets demonstrate that our method outperforms several state-of-the-art MF methods by 10% ∼ 28.4%. Our implementation is available at: https://github.com/ dingjingtao/View_enhanced_ALS.

Unbiased Learning to Rank

2021 ◽  
Vol 39 (2) ◽  
pp. 1-29
Author(s):  
Qingyao Ai ◽  
Tao Yang ◽  
Huazheng Wang ◽  
Jiaxin Mao
Keyword(s):  
Online Learning ◽  
Theoretical Foundation ◽  
Learning To Rank ◽  
Research Question ◽  
Parameters Estimation ◽  
User Interactions ◽  
Empirical Performance ◽  
Search Data ◽  
Two Sides ◽  
Important Research Question

How to obtain an unbiased ranking model by learning to rank with biased user feedback is an important research question for IR. Existing work on unbiased learning to rank (ULTR) can be broadly categorized into two groups—the studies on unbiased learning algorithms with logged data, namely, the offline unbiased learning, and the studies on unbiased parameters estimation with real-time user interactions, namely, the online learning to rank. While their definitions of unbiasness are different, these two types of ULTR algorithms share the same goal—to find the best models that rank documents based on their intrinsic relevance or utility. However, most studies on offline and online unbiased learning to rank are carried in parallel without detailed comparisons on their background theories and empirical performance. In this article, we formalize the task of unbiased learning to rank and show that existing algorithms for offline unbiased learning and online learning to rank are just the two sides of the same coin. We evaluate eight state-of-the-art ULTR algorithms and find that many of them can be used in both offline settings and online environments with or without minor modifications. Further, we analyze how different offline and online learning paradigms would affect the theoretical foundation and empirical effectiveness of each algorithm on both synthetic and real search data. Our findings provide important insights and guidelines for choosing and deploying ULTR algorithms in practice.

scholarly journals Dynamic Incentive-Aware Learning: Robust Pricing in Contextual Auctions

2020 ◽  
Author(s):  
Negin Golrezaei ◽  
Adel Javanmard ◽  
Vahab Mirrokni
Keyword(s):  
Decision Maker ◽  
Strategic Behavior ◽  
Learning Algorithm ◽  
Online Advertising ◽  
Ground Truth ◽  
Decision Makers ◽  
Effective Learning ◽  
Reserve Prices ◽  
Feedback Data ◽  
Over Time

In many practical settings, the decision makers have to learn their best actions by experimenting with possible options and collecting feedback (data) over time. It is often assumed that the collected data can be trusted as they reflect the ground truth. But this assumption is violated when the data are generated by strategic players. Consider online advertising market in which the ad exchange (decision maker) aims at learning the best reserve prices in the repeated auctions. In this setting, the data are advertisers’ submitted bids. Such data can be strategically corrupted by advertisers to trick the learning algorithm of the ad exchange to offer them lower reserve prices in the future auctions. In “Dynamic Incentive-Aware Learning: Robust Pricing in Contextual Auctions,” N. Golrezaei, A. Javanmard, and V. Mirrokni design effective learning algorithms with sublinear regret in such environments that are robust to the strategic behavior of the players.

scholarly journals Neural Matrix Factorization Recommendation for User Preference Prediction Based on Explicit and Implicit Feedback

2022 ◽  
Vol 2022 ◽  
pp. 1-12
Author(s):  
Huazhen Liu ◽  
Wei Wang ◽  
Yihan Zhang ◽  
Renqian Gu ◽  
Yaqi Hao
Keyword(s):  
Neural Network ◽  
Matrix Factorization ◽  
Recommendation System ◽  
Heterogeneous Data ◽  
User Preference ◽  
Personalized Recommendation ◽  
Implicit Feedback ◽  
Network Training ◽  
Feedback Data ◽  
Explicit Feedback

Explicit feedback and implicit feedback are two important types of heterogeneous data for constructing a recommendation system. The combination of the two can effectively improve the performance of the recommendation system. However, most of the current deep learning recommendation models fail to fully exploit the complementary advantages of two types of data combined and usually only use binary implicit feedback data. Thus, this paper proposes a neural matrix factorization recommendation algorithm (EINMF) based on explicit-implicit feedback. First, neural network is used to learn nonlinear feature of explicit-implicit feedback of user-item interaction. Second, combined with the traditional matrix factorization, explicit feedback is used to accurately reflect the explicit preference and the potential preferences of users to build a recommendation model; a new loss function is designed based on explicit-implicit feedback to obtain the best parameters through the neural network training to predict the preference of users for items; finally, according to prediction results, personalized recommendation list is pushed to the user. The feasibility, validity, and robustness are fully demonstrated in comparison with multiple baseline models on two real datasets.

Quantum algorithm design using dynamic learning

2008 ◽  
Vol 8 (1&2) ◽  
pp. 12-29
Author(s):  
E.C. Behrman ◽  
J.E. Steck ◽  
P. Kumar ◽  
K.A. Walsh
Keyword(s):  
Quantum Interference ◽  
Quantum Computer ◽  
Learning Algorithm ◽  
Algorithm Design ◽  
Quantum Algorithm ◽  
Initial Time ◽  
Dynamic Learning ◽  
Final Time ◽  
Quantum Neural Network ◽  
Qubit System

We present a dynamic learning paradigm for ``programming'' a general quantum computer. A learning algorithm is used to find the control parameters for a coupled qubit system, such that the system at an initial time evolves to a state in which a given measurement corresponds to the desired operation. This can be thought of as a quantum neural network. We first apply the method to a system of two coupled superconducting quantum interference devices (SQUIDs), and demonstrate learning of both the classical gates XOR and XNOR. Training of the phase produces a gate similar to the CNOT. Striking out for somewhat more interesting territory, we attempt learning of an entanglement witness for a two qubit system. Simulation shows a reasonably successful mapping of the entanglement at the initial time onto the correlation function at the final time for both pure and mixed states. For pure states this mapping requires knowledge of the phase relation between the two parts; however, given that knowledge, this method can be used to measure the entanglement of an otherwise unknown state. The method is easily extended to multiple qubits or to quNits.

scholarly journals A music recommendation algorithm based on clustering and latent factor model

2020 ◽  
Vol 309 ◽  
pp. 03009
Author(s):  
Yingjie Jin ◽  
Chunyan Han
Keyword(s):  
Collaborative Filtering ◽  
Large Scale ◽  
User Preference ◽  
Implicit Feedback ◽  
Music Recommendation ◽  
Recommendation Algorithm ◽  
Latent Factor ◽  
Latent Factor Models ◽  
Feedback Data ◽  
Preference Matrix

The collaborative filtering recommendation algorithm is a technique for predicting items that a user may be interested in based on user history preferences. In the recommendation process of music data, it is often difficult to score music and the display score data for music is less, resulting in data sparseness. Meanwhile, implicit feedback data is more widely distributed than display score data, and relatively easy to collect, but implicit feedback data training efficiency is relatively low, usually lacking negative feedback. In order to effectively solve the above problems, we propose a music recommendation algorithm combining clustering and latent factor models. First, the user-music play record data is processed to generate a user-music matrix. The data is then analyzed using a latent factor probability model on the resulting matrix to obtain a user preference matrix U and a musical feature matrix V. On this basis, we use two K- means algorithms to perform user clustering and music clustering on two matrices. Finally, for the user preference matrix and the commodity feature matrix that complete the clustering, a user-based collaborative filtering algorithm is used for prediction. The experimental results show that the algorithm can reduce the running cost of large-scale data and improve the recommendation effect.

Sign in / Sign up

Export Citation Format

Share Document